This invention relates to an optical disk on which data is recorded, a recording method for recording data on the optical disk, and an apparatus for recording and reproducing data on the optical disk.
Recently, DVD systems have been developed to meet the requirements for recording MPEG2 images on one surface of an optical disk having a diameter of 12 cm for two hours or more. Present DVD standards state that the memory capacity of the disk is 4.7 GB for one surface, the track density is 0.74 μm/track, and the linear density is 0.267 μm/bit. A DVD based on the above standards is referred to as a present DVD in this specification.
Reproduction of information recorded on an optical disk such as a DVD is performed by use of an optical head. In the optical head, a light beam emitted from an LD (laser diode) is focused on a pit series formed on the track of the optical disk via an object lens, a light beam reflected from the optical disk is concentrated on a photo-detector by use of a condensing lens and thus a reproduction signal can be attained. The reproduction signal from the photo-detector is input to a reproduction signal processing system and subjected to waveform equalization by an equalizer, and then data is decoded by use of a detector. According to DVD standards, the wavelength of the LD in the optical head is 0.65 μm and the numerical aperture of the object lens is 0.6.
As a higher definition system is required as an image source, it is required to further increase the memory capacity of the DVD in order to record and supply such image data. In order to satisfy the above requirement, the wavelength of the LD is reduced to attain a higher recording density expressed in terms of the wavelength, thus requiring further study on increasing the memory capacity using the PRML (Partial Response Maximum Likelihood) signal process.
In the conventional signal detection method, whether the recorded information is “0” or “1” is determined for each bit. Since the interval between pits or marks becomes shorter as the information recording density is increased, the influence of waveform interference by the adjacent information bit on the reproduction signal also increases.
In order to eliminate the influence of waveform interference, it is necessary to perform a signal process to emphasize the high frequency component of the response characteristic of the recording/reproduction channel and suppress the skirt portion of the response waveform extending to the adjacent bit to a low level.
However, since the noise component is also emphasized if the high frequency component of the response characteristic is emphasized, the number of errors caused by the emphasizing process consequentially increases. Therefore, it is difficult to significantly improve the recording density using the conventional signal processing system.
On the other hand, in the PRML signal processing system, a waveform interference amount between adjacent bits of the reproduction signal waveform is permitted in a range specified by the PR (Partial response) class. Since the reproduced waveform is influenced by the waveform interference from the adjacent bit and distorted, it becomes impossible to determine data by use of only one bit as in the conventional system.
However, since the waveform interference amount is limited to a specified value, signal power dispersed before and after an information bit can be efficiently utilized if a Maximum Likelihood detector using an ML (Maximum Likelihood) determination circuit for selecting data of the most likelihood among the sequence by taking portions before and after the waveform into consideration is used, and therefore, data can be detected at a relatively low error rate. A PR equalizer is used as an equalizing circuit for correcting a deviation of the reproduction waveform from the PR class.
In conventional level slice equalizers, the reproduced waveform is subjected to a waveform equalization process to set the intersecting point between the equalized waveform and a certain threshold level at the center of the window. More specifically, the high frequency component of the reproduced signal is amplified. In the detector, an intersecting point between the equalized waveform and the certain threshold level is detected and if the intersecting point is detected in the window, binary data “1” is output, and if it is not detected, binary data “0” is output. Then, by subjecting the binary data obtained after detection of the intersecting point to an NRZI (Non Return Zero Inverted) conversion process, the decoded data can be attained.
The intersecting point between the equalized waveform and the threshold level does not always exist at the center of the window due to the presence of noise. The standard deviation of intersecting point data standardized according to the window width is called “jitter” and is used as an evaluation standard for optical disks and drive units.
Now, a case wherein an optical disk whose recording density is made higher than present DVDs is reproduced by use of the same optical head as that of the present DVD is considered. If the track density is increased, a reproduction signal may contain a large amount of signal-degrading component called “crosstalk”. On the other hand, if the linear density is increased, the reproduced waveform becomes dull. As described before, in the equalizer, since the high frequency component of the reproducing signal is amplified, it is necessary to more strongly amplify the high frequency component when an input reproduced waveform becomes duller. As a result, the equalizer also amplifies the signal-degrading component. Thus, if the waveform slice system is used for the signal detection system, the signal-degrading component is increased irrespective of the method for increasing recording density, therefore data cannot be correctly decoded.
As the reproduction signal processing system used when the SNR (signal/noise ratio) of the reproduction signal is lowered, utilization of a PRML (Partial Response and Maximum Likelihood) system instead of the waveform slice system is studied. In the PRML system, the reproduced waveform is equalized into a waveform having a known correlation between identification points called a PR characteristic by use of the equalizer.
Further, the PRML signal processing system can attain a satisfactory (low) error rate in the case of high recording density. This system is a system for detecting data by use of the correlation between information bits while permitting waveform interference. Therefore, a delay occurs because a sample data series is stored into a bus memory as shown in FIG. 9 which will be described later.
In the process for detecting user data, generally, the above delay does not cause any problems. However, in the header field, since a readout sector number is fed back to the later operation, the influence becomes larger.
That is, in a data writing process, a sector number is read out from the header field, and if the number is determined as a to-be-accessed sector, the readout operation must be immediately interrupted, which affects the write operation. Further, if the number is not the object sector, the write operation is disabled since data will be destroyed. Therefore, if detection of the number necessary for the above determination is delayed, it becomes necessary to take a large GAP field of FIG. 10 which will be described later in order to take a sufficiently long time corresponding to the delay time and thus the format efficiency is degraded accordingly.
In the conventional header field, only one address mark AM for attaining byte synchronization is provided after VFO. The construction is satisfactory when a clock phase pull-in process is completely terminated in the VFO field, but if the phase pull-in process is not completely performed due to a disturbance, for example, data after this cannot be correctly read out.
Therefore, it is required to provide a device capable of suppressing the time delay caused when an address of the header field is read out, reducing the interval of a gap field required between the header field and the recording field and improving the format efficiency.
Further, it is required to provide a device capable of error-free readout of an address of the header field.
This invention can solve the above problems, suppress the time delay caused when an address of the header field is read out, reduce the interval of a gap field required between the header field and the recording field and improve the format efficiency.
Further, this invention can read out an address of the header field without error.